( 261 )

FOUNDATIONS OE THE SAXON CATHEDRALCHURCH AT ROCHESTER.

BY THE REV. GREYILE M. LIVETT,MliTOB OAH-QH- AND KREOEHTOB,

ITS the autumn of the year 1888 the Dean and Chapter ofBochester, acting under the advice of Mr. J. L. Pearson, K. A.,decided to proceed at once •with the underpinning of thewest front of the cathedral church, preparatory to itsrestoration. The work was put into the hands of Mr. JohnThompson, contractor, of Peterborough. In the course ofthe excavations some important discoveries were made, andit is the object of this paper to record them, and to shewtheir relation to the history of the church. 'The presentwest front, which is Norman, marked in blue outline in thelarger part of Plate I., was found to rest upon an earlierwall, also of Norman date, which had been demolished towithin about 2 feet 6 inches of its foundations, marked redin the Plan. In this paper, therefore, I shall call the exist-ing front Later-Norman work, and the remains of its pre-decessor I shall call Early-Norman. The Early-Normanwork, however, must not be confused with the still earlierNorman of Bishop Grundulf, though it belongs to histime. GrundulPs work will be called by his name, except-ing when the first Norman church is referred to as a whole,and then both works are to be included, since the EarlyNorman front was built to complete G-undulf's unfinishedchurch. The remains of the Early-Norman west front werenot the only interesting discovery we made. Runningunder and through its foundations, were discovered thefoundations, and portions of the walls, of a building of farearlier date than the earliest of the Norman works. Theseolder foundations underlie the northern half of the presentwest front, and are marked. black in the Plan. They doubtless

Archaeologia Cantiana Vol. 18 1889

262 FOUNDATIONS OE THE SAXON

form part of the Saxon church, the documentary evidence for•which. Mr. W. H. St. John Hope has collected,* and thediscovery of which Mr. J. T. Irvine, years ago, prophesiedwould be made, whenever the west front should be under-built.f

The following table of the dates of the bishops who willbe mentioned will clear the way for a description of thediscoveries:

1. JUSTUS (translated to Canterbury) 604— 6243. PATOIITTTS (olim Bishop of York) 633— 6444. YTHAMAB the Englishman 644— 6559. TOBIAS the Learned 693—726

29. G-UITDUI.:P (monk of Bee, and Lanfrano'schamberlain) 1077—1108

30. RAIES KE TUBBINE (translated to Can-terbury) 1108—1114

31. EBNTTKF (prior of Canterbury, and abbobo£ Peterborough) 1115—1124

32. JOHN HE CAOTEBBURY 1125—1137

I. THE SAXON CHURCH.

The underpinning, of the west front, was carried out insections. • To attempt to describe the piecemeal discovery ofthe earlier foundations would be tedious and uninteresting.The reasons which lead us to identify these foundations withthe Saxon church are threefold. In the first place, historytells us a church was built at Bochester in the year 604.Secondly, the character of the discovered masonry is whatone would expect to see in work of that period, and the planof the building could hardly belong to anything else than achurch. And, lastly, the ground on the south and east ofthese foundations is full of graves, which lie exactly parallelto the axis or line of orientation of the building erected onthe foundations.

* See his Notes on the Arehiteetwal 'History of Rochester Cathedral Church,and a communication to the Society of Antiquaries entitled Q-undulf's Tower atRochester, and the first Norman Cathedral Church there.

t MS. notes. Mr. Irvine was Clerk of the "Works to Sir G-. G-. Soott duringthe restoration of the cathedral which was carried out between the years 1871and 1878. He is now superintending the work at Peterborough for Mr. Pearson.I am much indebted to Mr. Irvine for tracings) and for information of what wasdiscovered at that time, over and above that supplied in his MS. notes.

CATHEDKAL CHURCH AT HOOHESTER. 263

1. In the year 604, seven years after Ms arrival inEngland, St. Augustine established the two sees of Londonand Bochester, and ordained Mellitus and Justus to fillthem. King JUthelbert built the church of St. Andrew inBochester, and endowed it with lands.* Beda tells us itwas built a fwidamentis (from the foundations): an expres-sion which would refer to a building of stone rather thanone of wood. Seventy years later, -when the Merciansinvaded Kent, the city was sacked and the church spoiled;but the actual fabric seems to have survived both this andlater invasions, for no statement to the contrary is found.This view is confirmed by the fact that the sites of thegraves of Paulinus and Ythamar, "who were buried in thechurch, were known up to the eleventh century. Gundulffound the Saxon Church almost a rain, built a new church, •and transferred into it the relics of Paulinus, whose graveup to that time had evidently not been disturbed. So muchfor the historical evidence, which certainly gives the Saxonchurch an unbroken existence from the seventh to theeleventh century.

2. The discoveries made indicate a building terminatingtowards the east in an apse, the width of the apse beingalmost as great as that of the building itself. A sleeper-wall lies along the chord of the apse. In the Plan, thefoundations are hatched, and denned -with a bounding lineonly where they were actually disclosed in -the excavations.The wall, where seen, is marked in solid black, and the con-jectured parts of it are cross-hatched. These foundationswere first struck, by the workmen, along the southern half ofthe sleeper-wall on the chord of the apse. They wereworked out to the bottom thereabouts, and probably a smallpart of the actual apse-wall disappeared in the process. Idid not arrive on the scene until this had been done; andthen I was told that the concrete was so hard that the ordi-

* See ffrowf ly 2Efhelbert, King of Kent, to St. Andrew's, Rochester, of landat Sauthgate, 2&J/pril 6CH, which occupies a prominent place in Mr, Walter deGray Birch's Cartulwium Saxoniautn. See also Thorpe's JRegistnm Zoffenselae chief additional authorities for the history are : for the early Saxon -period'me pooles. Hist, of the Yen. Bede; for the early ISformaa period Ernulfsleatus Zoffensis, and a V4fa ffundulfi, written by a monk of Rochester about1115, or a httle later, and printed by Wliarton in his Anglia Sacra.

FOUNDATIONS OF THE SAXON

nary picks were not stout enough to deal with it, and thatstrong iron chisels had been especially made for the purpose.At the bottom of the foundations, at this point, a large"sarsen" stone was found, embedded in the mortar: itis now in my garden. Then followed the discovery of thefoundations of the apse, inside the present building. Lateron, the junction of the apse with the north end of thesleeper-wall was disclosed. The upper part of the internalquoin had been removed, to mate way for a modern circularbrick drain.* Lastly, the junction of the apse with the eastend of the south wall of its nave was found. This was themost important "find" of all; and our thanks are due to theDean and Chapter for the leave they gave me to have twodays of extra digging to try to discover this junction. Its im-portance lies in the fact that it has afforded us the bestexample of the masonry of the actual walls. Even here thewall remained to a height of only about 1 ft. 8 in. above thefoundations. A slight sketch of the masonry may be seen inPlate II., No. 3. -The quoin consists of two ferruginous sand-stones, faced, and of large size; the angles are much worn away.A. suggestion of herring-bone work will be noticed, and alsothe use of an 11-inch Roman brick (of a drab colour). Thework is exceedingly irregular, and the joints large. Themortar is very hard^ and made of a sharp flint sand, with afew shells and some charcoal in it. Sketch No. 3 also shewsthe two courses of Roman brick which alone remained to markthe line of the apse on this side. The Roman bricks wereof different colours, drab, buff, and red, some broken, somewhole. There was also a portion of a flue-tile. All thesewere evidently old materials, used again. Portions ofthe apse-wall remained on the foundations elsewhere, asshewn in the Plan, but they consisted merely of one courseof long roughly-squared stones, some of tufa, others of Kentishrag. The walls were 2 feet 4 inches in thickness.

A section of the foundations is given in Plate II. (No. 6).The dimensions vary slightly, but the depth is about 4J feet,

* Bodies were found near the centre of the sleeper-wall, -which had appa-rently been partly worked away to receive them. It may he an error therefore,though it is convenient, to call this a sleeper-wall.

>' -̂ ?

»—*

i. quNDULF. I 2.. SARLY" NORMAN.' ' ~ 'Y . NORTH AISLC.lst.8AY.

v/4-

•5SVNBaB|MJ<£££J

C.

Q.

3. SSOTION ACROSS FOUNDATIONS OF WEST FRONTNCAR N,w. DOOBWAY,

fi 6 o

fr^cx^nf^WtfofVj

SARt-Y NOR.4,FOUNDATION DITOH

Ist.BAY.S.AISLS.

SAXON CAFUlY NOR.

5. CCNTf tAU DOOR,vyAY,WGST FRONT.)

8

exT.4- -*• INT.

V^LJ.

(DesTRpv^D

st....

cofJ e f^eje

FOUfJo/^l-IOf"!

^-.-_\6. SAXON

6ASTSRN APSS.

°r

,/ q

BtfSL\T6t\HoRMAfl-.T^r1FOUfJOAp 'IK

. . < . . .tnsrvs. el uftL.

J889-

FOUNDATIONS UND6RY. N.W. TOW6R. '

SCALE OF PSST AND (NCHeS

~ ' "

4. SARLY NORMAN WALLING

i. EARLY NORMAN BASES. SAM IAN V 'ARe .3. SAXON WALLING.

3. SAXON SCULPTURED STONC

2,. SARLY NORMAN WAULINq.

C.F.KELL.PHOrO-LITHO.S.FUPlNIVAL ST HOUBOBN.E.C.

5,6. MQULDINqS FROM LAT6R NORMAM FOUNDATION WORK. 8. UPCHUROH V e s 8 S L .

Rochester Catliedral, Plate 11, Detgils of discoveries at west end of Nave,ieJ83.

IO. FRACJM6NT9

CATHEDRAL CHURCH AT ROCHESTER. 265

and the width about 4 feet. On the inside the vertical wallof foundation-work is regular and fairly smooth; on theoutside it is very rough and uneven. Where they passunder the north turret (from the inside) the foundationsare wider at the top than at the bottom, as shewn in thesection; on the opposite side the reverse is the case. Thematerial is the same hard mortar already described, withsmall pebbles in it, and blocks of Kentish rag laid in, with-out attempt, it seems, at regular coursing. They rest, gene-rally speaking, on the maiden soil, which is found aboutseven feet below the present level of the road in front of thecathedral. But the Saxon builders do not seem to havebeen very particular in this matter, for the eastern part ofthe apse-foundations, where they emerge from, under thenorth-west turret on the inside, lies over a pit or ditch ofsoft black earth. The eastern side of this pit, or whateverit was, was very plainly seen in the excavation made in thiscorner. It ran about north and south. !From the blackmould right underneath the foundation-work, I extracteda piece .of very rough pottery, a piece of Koman brick,oyster-shells, bones, and charcoal. It may be that thisunsound bottom caused a settlement, which would accountfor a large crack that we saw in the foundation-concrete,extending right down to the bottom, just at the point wherethe Early-Norman respond-f oundations run up to it. Thisfault must have occurred before the church was demo-lished in Norman times, and is a confirmation of the truthof the record -which tells us that Gimdulf found the churchalmost ruined from, old age, pcene vetustate dvrutam.

3. A description of the graves and burial-ground remainsto be given. Mr. Irvine tells me he found that the ditch ofthe foundations (Norman), of the north aisle-wall of thepre-sent church, cut through a Saxon burial-ground along itswhole length, and that it did not reach the bottom of theground used for that purpose. In the recent excavations wefound a somewhat remarkable Saxon grave, near the north-east corner of the north turret. It is marked in the Plan,It lay between 5 and 6 feet below'the level of the roadwaj

266 . FOUNDATIONS OF THE SAXON

hard by.* The sides were formed of large stones, three onone side aud four on the other, set close together in a looseand soft mortar, f The stones did not extend the wholelength of the grave, and there was none at the head or foot.The tops of the stones were on a level with one another, andthere was mortar on them, suggesting the existence origin-ally of a covering slab. It is possible that when this wasremoved some of the supports were removed also.J Thewidth of the grave inside was 1 foot 6 inches, and its depth1 foot, The bottom was formed of a bed of mortar, onwhich there was a thick coating of brick-dust. On this thebody lay, covered (when found) with mould. That this wasa Christian burial is evident from the absence of any knives,weapons, or personal ornaments, usually found in thegraves of pagan Saxons. Its character, its depth below thesurface, its position in relation to the apse, all point to thelikelihood of its having been one of the earliest of the Saxonburials.§

Another body was found under the steps inside the cen-tral west door, on the right hand side. This was one of thefirst discovered, and I did not see it. The fact, however, isworth recording.

The Saxon burial-ground was again tapped, when searchwas being made for old foundations, in front of the southernjamb of the central doorway. The underpinning of thispart of the west front had been completed some time before,and the foreman distinctly remembered cutting through somemasonry at a low level thereabouts, so I determined to use part

* The surface of the graveyard, called in medieval times Green ClmrohHaw, is now much higher than the road.

f At the foot of the grave there was much of this mortar, and in it twopieces of white plaster and some bits of a much harder mortar.

t The faces of these stones ate not rectangular: they taper slightly, like thevoussoirs of an arch. The hacks are roughly wedge-shaped, the line of the ridgerunning at right angles to the tapering sides. All this points to the fact thatin this grave they were not serving the purpose for which they were made.

§ I saw the Saxon grave in position, but I did not see the removal of thebody, nor of some black mould which lay beneath the grave, aud which con-tained some hones and shreds of rough black ware. Examining the spot after-wards, I clearly saw the maiden soil undisturbed at the side of the hole whichhad held the black mould and ware. At the bottom of the hole there were signsof burnt wood in a damp pasty condition., ,

CATHEDRAL CHURCH AT ROCHESTER. 267

of the two days of extra digging, allowed by the Dean andChapter, in trying to rediscover it. The hole was dug so asto escape the concrete recently laid down. From 3| to 4 feetbelow the paving line we reached a bed of very rough mortar,varying in colour, bub chiefly yellow, containing bits oftufa, of Roman brick, and Eoman mortar. It "was beddedupon a layer of chalk and flints. Charcoal was found plen-tifully strewn over it. On this bed rested bodies lookingeastwards, and lying close up to the Saxon foundations.There were no weapons or personal ornaments.*

* Underneath the bed of mortar on which the bodies lay there was a stratumof mould and rubbish, rather hard towards the bottom, and containing bits ofRoman ware. It was from this level we got the Sa,mian fragment which lafigured in Plate II. (No. 7). It will be noticed that the potter's name,oimsA .. ., is worked into the pattern down the Bide of the vessel. Prom theblack moist mould at the bottom, immediately above the maiden soil, which liesabout 7 feet below the paving line, were taken up (in my absence) the pieces ofa large urn of coarse ware, much broken by the workman's pick, but I do not thinkit was perfect. Among the fragments were the necks of at least three otherurns, one fire-marked. The bottom of the large urn came out fairly complete,and in it there was a great number of minute bones (? Aryicola), together withthe pieces of a small vessel of Upchurch ware of good quality. It had evidentlybeen broken by the workman, and the foreman, Mr. litzjohn, who took greatinterest in these matters, worked away at the pieces, and contrived to restore thevessel almost complete. An outline sketch of it is given in Plate II. (No. 8). Itstands 2f inches high, and measures 5 inches across the top.

In the level which I have just described we found the masonry we werelooking for. Some of the bodies rested partly upon it. We first hit upon anexternal quoin, the faces of which happened to be nearly right with the direc-tion of the Saxon church. The outline of the -quoin is marked in the Plan,Its southern face disappeared into the modern concrete; its western facewe followed up until it nearly reached the foundations of the south wall ofthe Saxon church. I could not interpret the evidence of the junction of thetwo works: it was very rough; nor could we ascertain the thickness of ournew wall. Its foundations, measuring only 18 inches in depth, had been sunkthrough the black mould, and rested on maiden soil. They consisted of flints andrag-stones set in a yellow sea-shore mortar. Portions of the wall remained onthem. Its mortar'was harder and stronger. At the quoin two Roman brickswere _ set in the foundations with wide mortar-joints, and above them thereremained one of the squared quoin-stones of tufa. The rubble wall was madeof flints and rough blocks of tufa, one of which shewed a small portion of facingof a fine, compact, white plaster. This will be preserved in the crypt. It hasthe charcoal adhering to it on another side. The quoin was left in situ.

In the course of the excavations many pieces of pottery of various kinds werefound; also a boar's tusk and the core of horn of the 5os longifrons; andmany broken Roman bricks, flange tiles, and flue tiles, A very interesting frag-ment, measuring only 6 inches by 2J to 3J, of a stone 3 inches thick, came fromsomewhere under the nave floor. It is rudely sculptured on both sides in whatMr. St. John Hope palls " strap pattern," and shews plain signs of red andbrown colour, which is easily rubbed off. See Sketch No. 10. The lines on thebottom look like part o£ an inscription. It will be preserved with the otherIragments m the crypt. Another stone of interest was found built into the.Later-.Norman plinth: it is a portion of a Saxon sepulchral slab, and shews aninterlaced, pattern and the hind-quarters of an animal sculptured on it (No 9)

268 FOUNDATIONS OP THE SAXON

This is all I have to say now about the building and theburial-ground. Mr. Chas. Eoach, Smith kindly came tolook at the masonry at the junction of the apse with thesouth wall, and he gave his opinion that it was "not Eomanwork, Ibut close on Roman times." Mr. St. John Hope camedown from London, and saw part of the apse and of thesleeper-wall, and he has since expressed to me a strongopinion that the remains formed part of the church ofKing JEthelbert.* It is the only conclusion we can cometo; and if the discovery from its limited nature is not veryimportant, it is exceedingly interesting as giving us someknowledge at least of one of the very earliest churches builtin our country. It was doubtless built by Saxon hands,though it shews some Eoman influence, either traditional ordirect. The materials probably came from destroyed Eomanbuildings; the foundations of such have at various times beenfound in two or three spots riot far distant, at a low level inthe ground. What the rest of the ground-plan of thechurch was, it would be impossible to say, with any certainty,from the existing data. The substantial foundations alongthe' chord of the apse must have supported something, and Iam inclined to think there may have been a chancel arch, asat Brixwortk, but I have not ventured to indicate it in thePlan. At Brixworth, too, the aisles commenced some thirtyfeet west of the apse, and the apparent want of aisles atEochester might be accounted for by a similar arrangement.The foundations run westward under the strip of, grave-yardin front of the cathedral church. Perhaps a trench may atsome future date be dug here and there in that ground, inorder that the whole of the ground-plan may be recovered.It has been suggested that the outlines of the apse and wallsshould be marked, in some way, on the floor inside and on thepavement outside the west front, and that the followingwords from Beda should be inscribed, within the lines, on thefloor inside: A.D. 604 BEX .ZEDILBERCT ECCLESIAM BEATIANDREW APOSTOLI KECIT.

* Mr. A. A. Arnold, Mr. Q-. M. Arnold, and Mr. George Payne also sawportions of the work at different times.

CATHEDRAL CHURCH AT ROCHESTER. 269

IE. THE EAKLY-NOKMAN WEST FBOHT.

1. The first thing the great Norman architect Gundulfdid, in the city of Bochester, was to build a massive squaretower, at a distance of 150 feet from the Saxon church east-ward. No one knows for certain what purpose the towerwas meant to serve in the first instance, but seventy yearslater it was being used as a campanile. It is shewn in solidblack in the small Plan of Plate I. Gundulf had beenbishop only a few years when he replaced the four secularcanons who remained in the impoverished college by twenty.Benedictine monks, and set about building a new and largerchurch for them.* He was enabled to do this by the re-covery, with the king's help, of certain lands which Odo ofBayeux had usurped, " After a brief interval, therefore, whenthe old church had been demolished, the new one was begun;a circuit of offices [for the monks] was conveniently arranged;and the whole work finished within a few years by themunificent help of Lanf rano." So, with certain inaccuracies,wrote the monk of Eochester thirty or forty years after-wards ; and he adds that when all things were ready Lan-franc went in solemn procession with the monks and clerksto the grave of Paulinus in the old church, and translatedthence his sacred relics into the new church. This transla-tion took place in the year 1087, which is doubtless the dateof the close of Gundulf's building operations. Two otherfacts are equally certain: (i) the Saxon church was not de-molished before the new one was begun; and (ii) Gundulf hadnot finished his church. So much as he did accomplish isshewn in solid black in the small Plan. Now arise the ques-tions : (i) Why did Gundulf complete the south aisle wall andthe great arcade on the south side of the nave, and stop half-way on the north side ? (ii) When was the work taken upagain and finished? The answer to these questions may be

. * .It -will he noticed that tlie church orientates nearly due south-east Thisis owing to the fact that Gundulf was in a strait for want of space : he could notgo iurtner north (towards the east) on account of Ms tower, so he arranged thatthe tower should fall into the corner, between his north transept and choir aislewans, and He could not go further south (towards the west), or he would nothave had room for his circuit of offices between the nave and the south wall of

270 - FOUNDATIONS OP THE SAXON

drawn from tlie fact that previous to the year 1423, -when thechurch of St. Nicholas was built upon a piece of land inGreen Church Haw, given by the monks for the purpose,the parishioners of St. Nicholas worshipped in the nave ofthe cathedral church. There is evidence in the RegistrumRoffense that the altar of St. Nicholas, " parochial in thechurch of St. Andrew," existed towards the end of Gu.ndulf'sepiscopacy. Doubtless this was only a fulfilment of Gun-dulf 5s original intention, that the nave or a part of it shouldbe assigned to the parishioners, though such intention is notexpressly recorded. This being the case it seems likelyalso, as Mr. St. John Hope has already pointed out, thatGundulf was content to build the monks' part only (exceptthat he was obliged to complete the soxrbh aisle in order tohave a wall against which to place the monks' cloister), andthat he left it to the parishioners to build their part.*The Early-Norman foundations we have discovered, belongto a building which is quite distinct from that of Gundulf jbut they cannot be more than a few years later in date.They are doubtless the work of the parishioners of St.Nicholas, undertaken, say, between the years 1095 and 1100.

Some twenty to twenty-five years later, when the Normanstyle had lost its early rude and plain character, thefirst Norman church was enriched, and partly rebuilt on theold lines. This work was probably begun by the greatbuilder Ernulf, and completed by his successor John deCanterbury,^ for the church was dedicated in the year 1180on Ascension Day. The Later-Norman front, which is builton the remains of its predecessor, was no doubt the latestwork of this partial rebuilding of the first Norman church;a rebuilding which might be called the second Normanchurch. With regard to the Saxon church, it probably

* The first site of the altar of St. Nicholas is not known. Early in the four-teenth century, it was sub yulpito, at the east end of the nave. Mr. St. JohnHope reminds me that G-undulf must have built a certain amount west of thecrossing to carry the thrusts of the crossing arches. Of course it is possible thathe may have originally meant to build the whole church at once, and that hestopped, for want o£ funds, just the later Gothic builders stopped. It is clearthat his flan and design were as plain and inexpensive as possible. !Por furtherinformation on the subject, and for a full account of the altar of St. Nicholas, Imust refer the reader to Mr, St. John Hope's Notes.

CATHEDRAL CHURCH AT ROCHESTER. 271

remained standing for the use of the parishioners until thebuilding of the Early-Norman front made it necessary todemolish the eastern parts of it. The western part, how-ever, was standing after that, for the remains of BishopYthamar, who, according to Weever, had been buried in thenave of the Saxon church, were not removed until the epis-copate of John de Canterbury. This removal probablymarks the final destruction of the old church.

2. In laying down the Early-Norman wall and founda-tions in the Plan, I have treated them in the same way as Itreated the Saxon work. I did not see the excavations onthe south side of the central doorway ; sufficient signs ofthe position of the quoins on. the outside remained after theholes had been filled up. On the inside, however, this wasnot the case, and, as no notes or drawings are forthcoming,I am obliged to assume that the Plan, corresponds with thatof the north side except in one particular. It is said thatsome barrel-loads of loose chalk were taken up from underthe respond of the great arcade. As Gundulf used loosechalk wherewith to fill his foundation-ditches, it seemslikely that these came from the end of the sleeper-wall of hisarcade. I have therefore indicated such a Gundulf sleeper-wall in the Plan by black dotted lines. It is also said thatunder the north side of the same respond there was a deepsinking of masonry. Not having sufficient data I could notindicate its position. It is possibly some Eoman foundation-work made use of by G-undnlf in connection with the sleeper-wall.*

The longitudinal section, No. 4, Plate II., shews thejunction of the two works in. the foundation-ditch under thesouth aisle wall in the westernmost bay. Gundulf 's ditchwas not much more than three feet deep, and did not reachmaiden soil. About one-third of it was filled with loosegravel, and then a quantity of chalk in small lumps, such as

* Foundations of Roman walls were found by Mr. Irvine in front of theaide watt. _ Mr. Koaoh Smith, too, tells me he has seen thef oundSs of awall running diagonally across Boley Hill, near the Oounlgr ]Krt»toThere were doubtless several Roman buildings within the walls ol the citv tart

~

272 FOUNDATIONS OF THE SAXON

could be dug -with an ordinary garden spade, were thrownin.*

The Early-Norman builders went down deeper, andreached the reddish earthy river-sand, containing flints,which constitutes the maiden soil; and they adopted a newmethod of filling their ditch. First, a double layer of flintswas laid in and covered with sand mixed with, a little lime.Then the ditch was filled with four layers of chalk and

' sand. This, Mr. Irvine remarks, was evidently meant to bea great improvement on the old method, but the builderswere probably using up materials left by Gimdulf, for whenthe ditch was opened up again, by Mr. Thompson's men,under the west front, in the recess near the corner turret,they found the use of chalk abandoned, and flints usedthroughout in similar layers, and set in mortar and redmould alternately. The depth of this work varied from 3£to 4 feet, and the thickness of the several layers variedconsiderably. The newer method was followed all along thewest front, f A still further improvement was noticed underthe north aisle wall in the first bay, where I had an oppor-tunity of making a close examination of the foundations.The Early Normans widened their ditch here, and made itssouthern face run across to join the apse-foundations, asshewn by the dotted line in the Plan, 'Plate I. The plan ofoperations here (see Section 2) had been to lay down a bedof mortar some three inches in thickness, and on it to throw alayer of flints. This in turn was covered with red sandymould, and the whole was rammed, and that so effectuallythat we found it very hard work to remove the flints withthe pick. All this had been done four times, and then, atthe top, came a layer of flints and mortar without mould,on which the foundation footing appeared. The mortarwas very loose, and contained pieces of a fine white plasterin considerable quantities—both wall and floor-plaster, appa-rently—aa well as a smaller number of fragments of thin

* On the chalk two courses of stone foundation, 11 inches <3eep3 were laid.These have disappeared from the south aisle wall (exterior), fcut Section 1 shewsthem as they were found by Mr. Irvine under the north aislo wall.

t See Section No; 3 in Plate II.

CATHEDRA! CHURCH AT ROCHESTER. 273

Boman stucco. Among the flints I found "what appeared tobe a portion of a Roman quern-stone.

I have described the contents of the Early-Norman ditchsomewhat at length, first, "because it indicates the progressof the work from the south side round to the north ; and,secondly, because it proves beyond all doubt that it wasquite distinct in point of time from G-undulf's work. Therubble walling is also distinctly of later character, thoughone could scarcely push its date into the twelfth century,Two bits of the wall are sketched in Plate II. (Nos. 2 and 4).The joints, filled with a fine yellow sand mortar, are wideand irregular. Otherwise there seems to be nothing callingfor remark in the masonry, except • it. be the use of theirbonding courses. In one of the examples this bondingcourse is slightly, but only slightly, suggestive of herring--bone work. The earlier rubble walling of Gundulf hasmuch more strongly marked features. An example to beseen in the south aisle (interior) shews most decided herring-bone work, in addition to the courses of thin and also ofbigger stones which occur in the Early-Norman work.* Asecond example of G-undulf 's may be seen in a portion ofthe north aisle wall (exterior), where the courses are muchthe same size, and all consist of stones set herring-bone-wise,with a larger stone set straight here and there only.

If these differences shew a distinction of date, there areenough points of likeness to bring the Early-Norman workclose to that of Gundulf. In both cases the walls rest on astone foundation (generally of two courses) of Kentish rag,which is only just wide enough to carry comfortably thepilaster buttresses. In both cases, walls and buttresses risefrom the footings without plinths j and in both cases tufais used for all the quoins, and Kentish rag for the faces ofthe rough walls and footings, with flints inside. The mea-rsurements correspond. The footing of the Early-Normanwork is generally about 1 foot 4 inches in depth and about afoot wider than the wall on each side. Sometimes a third

wimX &*•?* Gu^dUm 's WOS %?«* s^JlaF to ttat of the ^me architect'walling in St. Leonard's Tower, Mailing, which is figured in Parker's Introduction to Q-otMe Architecture.

274 FOUNDATIONS O3? THE SAXON

and thincourse is found "between the two larger courses, asmay be seen in Section 2. The Plan shews that at the endof the aisles the wall of the front is much thinner than else-where, and that the footing is correspondingly wider on theinside, and consists (as may be seen in Section 3) of a singlecourse of stones of great size, yery rough, the joints of whichcontain flints. The projection of the buttresses is 7-| or 8inches. That of the clasping buttresses of the corner pinna-cles may have been slightly greater. They were not exposedat either end, but I conclude that the supports of the cornerpinnacles of the west front were treated in this way (asshewn in the Plan), because it was the treatment which Mr.Irvine found adopted in the south transept of the firstNorman, church. The thinness of the wall at the end of theaisle would point to the same conclusion.

The ground-plan indicates a very plain front. Perhapsthe most interesting of: all our discoveries connected with itis that of the bases of the jamb-shafts of the central door-way, which was wider than its Later-Norman successor, butnot so deeply recessed. There are two bases remaining oneach side. The material is tufa. Those on the south side, byan arrangement of the new plinths,' will remain, not alwaysopen to view, but accessible. Including the mortar-bedthey stand about 7£ inches high. Base and quoin areworked together in one stone in:.'the ordinary way. Theplinths are square, the mouldings almost plain, and workedto a shallow central keel, suggestive' of the double-ogeemoulding so common in the perpendicular style, but reversed.The bases on the north side have of necessity been coveredagain. They are figured, however, in Plate II. (3STo. 1). Oneof them repeats the moulding just described, and. the othershews two plain rounds. The mortar-joints are large, vary-ing in width from 1 to 1̂ inches. The bases stand uponthe platform which carries the door-step. The bases of theshafts of the innermost order of the arch have disappeared,but the outlines of their mortar-beds .could be traced on thedoor-step on each side, so that they must have stood higherthan the others. The platform which carries the bases andthe step is 18 inches wide, and about 20 inches above the

CATHEDRAL CHURCH AT "ROCHESTER. 275

foundation-footing. TMs fact alone "would tell us that someof the Early-Norman -wall was underground from the first.Moreover, it is impossible that the surface of the groundshould have risen two feet and more in twenty-five or thirtyyears — the interval between the building of the early andlater fronts.*

A glance at Section 5, in Plate II., shews that the near-ness of the remains of the Saxon wall to the Early-Normandoorway, and the height of those remains in the ground,together preclude all possibility of the surface of the groundbeing so low as the Early-Norman footing. Moreover, wefound a pathway running right up to, and on a level with,the top of the door-step, which pathway, from its very posi-tion, must have been older than the Later-Norman work,and almost certainly coeval with the Early-Norman door-step. Its material, too, looked very much like that used inthe Early-Norman foundations. It consisted of two layersof plaster, with flints and red sandy mould between, thewhole being about 8 inches thick. I unfortunately failedto see either of the bases actually cleared, and so canonly suppose that the path sloped down on each side toallow the bases to be seen. On the interior, the originalwall-plaster, a very firm and hard white plaster, was foundon the Early-Norman wall running right down to thefooting, so there must have "been a considerable descent intothe nave of the church. This . awkward arrangement wasperpetuated and exaggerated by the Later-Norman builders.

We found no signs of the respond of the Early-Normanarcade on the north side, "but . its foundations remain andwere used to carry the Later-Norman respond. They runeastwards, and abut upon the foundations of the Saxonapse, indicating the line of the sleeper-wall of the arcade.I have ventured to shew this sleeper-wall beyond the apse" conjecturally." It is doubtful, however, whether the arcadeitself was ever raised, for no signs of it have yet been foundin the Later-Norman arcade, while on the south side the

lu/ ?kere is reason, however, to believe that it was in this interval that theti«?i,,w ? f °u*d

abeiween the west front (northern part) and the remains of

-toe half demolished Saxon church was used as a burial-ground.

0! 2

276 FOUNDATIONS OP THE SAXON

Later-Norman arcade undoubtedly is in the main Gundulf's•work.*- In fact, it is a question whether the EarlyNormans did after all quite finish their church. Perhapsfunds failed, and they were content for a time with sometemporary structure on the north side to support a tem-porary roof. It is significant that on the Early-Normanfooting of the north aisle wall, as shewn in Section 2, -thereremain only two courses of the Early Norman walling,not plastered) and that between them and the overlying coursethere is at least an inch of mould. This overlying course isundoubtedly Later-Norman work. It is of tufa, and has inone spot some of the original plaster adhering to it. Itsmortar is grey and shelly, and quite different from the yellowsand mortar used by the Early Normans in their walling. Thework above this is modern. The question thus raised seemsto be the only one of any real difficulty that has cropped up.It is not of any great importance to us now, but it should beborne in mind in future researches.

HI. It has already been said that the Later-Norman frontwas the closing work of what was practically a rebuildingof the first Norman church. It has suffered considerablyat the hands of restorers. The north pinnacle was rebuiltin its present odd octagonal shape in the sixteenth century.The north turret was rebuilt and finished off with a battle-ment at about half its original height in the worst style ofthe middle of the last century. At the same time, probably,the bases of the central doorway and the double plinthsthroughout were restored away, and a single plinth sub-stituted, leaving only a few of the original stones.f A little

* This was discovered by Mr. Irvine. The lower orders of the arches, andthe upper orders also on the aisle side, are all of tufa, now plastered over.

t It is this single plinth that is shewn in the Plan in blue. At the south-west corner the double plinth remained as a guide to the architect, and it hasjust been restored throughout. The bases, too, of the central doorway have beenreplaced. I was at first led to believe that the repairs which destroyed the basesand double plinth were made by Mr. Cottingham in 1826, but a list of Mr, Cot-tingham's work, which Mr. A. A. Arnold, the Chapter-olerk, handed to me, andof which I found a duplicate among some papers relating to the fabric whichthe 3Dean kindly allowed me to look through, makes no mention of any suchrepairs. The mortar used is most distinctive—a very tough, white mortar withgreen vesicles in it—and was found to have been used also for the two courses of

CATHEDRAL CHURCH AT ROCHESTER 277

later it was evidently found that the south, turret wasunsafe, and so little did the guardians of the fabric of thatday appreciate their splendid inheritance from the past, ortheir grave responsibility to tie future, that they adoptedthe ready cure of lopping off the top of it.* The wholefront, except perhaps the beautiful central doorway, is nowin a perilously dangerous condition, and the hideous shorescannot be taken down before it has been wholly restored.The whole of the face of the wall up to a considerable heighthas broken away from, the rubble core, and craclrs are deve-loping themselves in great numbers. The foundations werefound in a most unsatisfactory state. The wall overlaps theEarly-Norman wall on which it is built by as much as from34 to 4 feet along the southern part of the exterior. TheLatei<Norman builders, to widen the foundations, simplydug a trench along the earlier wall and foundations, but byno means to their full depth, and filled it with material ofthe same character as that of their predecessors, but lesscompact and serviceable.f The face of the wall aboveis very thin, seldom more than six inches, and scarcelybonded into the wall at all. One only wonders how anypart of it has stood so long. The careless way in which themasons of that day put up their work is well illustrated bySection 3, which shews how they placed the materials oftheir walling on the footing, and against the face of theearlier work, without attempting to bond them in. Under

foundation-work uuder the north turret (see Section 7), and for the steps insidethe small west door. Moreover, the tooling of the faced stone points to thelast century. So that I have now to come to the conclusion that these repairs,traces of which may still be seen, are of the same date as the north turret j andthis being the case, for the words, " Cottingham's plinths," in Sections 3 and 5,we must read, " eighteenth century plinths," and for "Cottingham's repairs" inSection 7, we must read, "and foundations." Cottingham's repairs at the westfront consisted in taking down and rebuilding in Bath stone the great windowand the battlements'above, and of partially repairing the stone-work of the twocorner turrets. Cottingham's repaivs awe and elsewhere cost nearly £10,000.

* A great deal has been done in the present century (more than £30,000 hasbeen expended by the Dean and Chapter since 1840), and much more remains to bedone to make amends for the apathy and neglect of past ages, and to put thefabric into a condition o£ safety. ,

t The material contained more Kentish rag than the Early Norman, evi-dently obtained, from, the destruction of the earlier wall. Some fragments o£apparently unused Norman mouldings were found in this foundation-workOne fragment shewed a triple nebule moulding, and u second shewed aoompony. They are sketched in >late II., Nos. 5 and 6.

278 WEST END OI1 ROCHESTER CATHEDRAL.

the north-west turret we found a stronger foundation thanelsewhere, but even there it did not reach the solid ground.*It consisted of great blocks of tufa and rag-stone, takenevidently from the destroyed Early-Norman front, and tallieswith the mass of masonry which Mr. Irvine found in frontof the north aisle wall in the first bay, and proves the cor-rectness of his inference that the Later-Norman architectmeant to flank his front with towers—a design which heevidently abandoned later on. Mr. Irvine found no such'preparation made on the south side.

It is perhaps worthy of remark that the walls of both theearly and later fronts taper towards the north. The earlybuilders worked by "rale of brow." The Later Normansused Caen-stone for all their facings and mouldings; andthey used up all the tufa which their predecessors leftbehind them, but in no case did they use it where it wouldbe seen. The plaster floor of the Later-Norman nave wascut through in several places during the excavations. Amoveable slab has been laid down in the present floor nearthe respond of the north arcade, where the old .floor may beseen seven inches below and running up to the bottom ofthe plinth of the respond. The plaster floor was veryuneven. It was found upon the Early-Norman footing,inside the north-west doorway (as shewn in Section 3), fromwhence it ran up on to the foundations of the apse, over atriangular bit of foundation work inserted by the LaterNormans into the corner, at the junction of the said footingand the apse, to support the turret.f Where there was nostone-work for it to rest upon, the soil was prepared by alayer of flints, on which the plaster, full of cockle-shells, waslaid. It formed a hard and.durable floor. Mr. Irvine foundsimilar floors under the choir and transept-crossing.

So my task comes to an end. It only remains for me tothank Mr. W. H. St. John Hope and Mr. Irvine for informa-tion and advice, and Mr. Thompson and his foreman forthe kind facilities they afforded me.

* See the interesting Section, No. 7, shewing works of three periods j butof. note f, P- 2^6.

- f The outline of this bit of foundation work is shewn by a three-dotted linein the Plan.

Kent Archaeological Society is a registered charity number 223382© Kent Archaeological Society